path: root/Master/texmf-dist/tex/latex/frankenstein/titles.sty

%^^A* legal notices
% \iffalse
%
%   This program is part of the Frankenstein bundle for LaTeX.  
%
%   Copyright 1995-2001 Matt Swift <swift@alum.mit.edu>.
%
%   This file contains both the code and documentation for the
%   titles LaTeX package.  It will work ONLY if it is placed in a
%   proper directory.  Files called README, INSTALL, titles.tex
%   and titles.ins should have also been distributed to you
%   with this file.  See them for more information on how to typeset
%   the documentation with LaTeX and how to generate a version of this
%   file that will work faster than this one.
%
%   This program is free software; you may redistribute it and/or
%   modify it under the conditions of the LaTeX Project Public
%   License, either version 1.2 or (at your option) any later version.
%   The latest version of this license is in
%   http://www.latex-project.org/lppl.txt, and version 1.2 or later is
%   part of all distributions of LaTeX version 1999/12/01 or later.
%
%   This program is distributed in the hope that it will be useful,
%   but without any warranty; without even the implied warranty of
%   merchantability or fitness for a particular purpose.  See the
%   LaTeX Project Public License for more details.
%
% \fi  
%
%^^A* checks
%
%^^A NOTE:  The character table, with two %'s, will get written to all files.
%% \CharacterTable
%%  {Upper-case    \A\B\C\D\E\F\G\H\I\J\K\L\M\N\O\P\Q\R\S\T\U\V\W\X\Y\Z
%%   Lower-case    \a\b\c\d\e\f\g\h\i\j\k\l\m\n\o\p\q\r\s\t\u\v\w\x\y\z
%%   Digits        \0\1\2\3\4\5\6\7\8\9
%%   Exclamation   \!     Double quote  \"     Hash (number) \#
%%   Dollar        \$     Percent       \%     Ampersand     \&
%%   Acute accent  \'     Left paren    \(     Right paren   \)
%%   Asterisk      \*     Plus          \+     Comma         \,
%%   Minus         \-     Point         \.     Solidus       \/
%%   Colon         \:     Semicolon     \;     Less than     \<
%%   Equals        \=     Greater than  \>     Question mark \?
%%   Commercial at \@     Left bracket  \[     Backslash     \\
%%   Right bracket \]     Circumflex    \^     Underscore    \_
%%   Grave accent  \`     Left brace    \{     Vertical bar  \|
%%   Right brace   \}     Tilde         \~}
%
% \CheckSum{783}
%
%^^A** abstract
% \begin{abstract}
%   The \package{titles} package defines macros that typeset the titles of
%   books, journals, etc. and handle following spacing and punctuation
%   intelligently, based on context.  These are useful for bibliographic
%   databases, for example.  Also defined is other markup like \cs\word,
%   \cs\defn, \cs\phrase, etc.
% \end{abstract}
% \tableofcontents
%
% \part{Discussion}
%
% \section{Options}
%
% There are two package options, \option{british} and \option{american}, the default
% is \option{american}.  They select the conventional way to use quotation marks:
% British style is use single quotes, and do not suck following period or comma
% inside; American style is to use double quotes and to suck following period
% or comma inside.
%
% \section{Words and phrases}
%
% \DescribeMacro\word
%   Typeset a word or phrase referred to as a noun with \cs\word\marg{word}.
%   The argument is not expected to contain punctuation.
%
%   \begin{bothexample}
%     \word{Elephant} is such a silly word.
%   \end{bothexample}
%
% \DescribeMacro\phrase
%   Typeset a phrase used as a noun rather than direct quotation with
%   \cs\phrase\marg{phrase}.  The argument might well have punctuation,
%   including final punctuation, which should not be considered to be
%   punctuation of the containing sentence.
%
%   \begin{bothexample}
%     The sentence \phrase{And stop calling me Shirley!} occurs
%     twenty-seven times.
%   \end{bothexample}
%
% \DescribeMacro{\foreign}
%   Typeset a foreign word or phrase with \cs\foreign\marg{foreign text}.
%   
%   \begin{bothexample}
%     I couldn't think of the \foreign{mot juste} at the time.
%   \end{bothexample}
%
% \DescribeMacro{\foreignword}
%   Typeset a foreign word or phrase referred to as a noun with
%   \cs\foreignword\marg{foreign word}.
%
%   \begin{bothexample}
%     Only later did I realize that the right word was
%     \foreignword{bouffon}.
%   \end{bothexample}
%
%   \caveat {Notice that writing \code{\foreign{\word{text}}} or vice versa is
%   not necessarily going to do the right thing.  Suppose \cs\foreign and
%   \cs\word were both set to \cs\textitswitch (which are in fact the default
%   settings below).  Then \code{\foreign{\word{text}}} is going to cancel out
%   and look just like the surrounding text.  This is not the most intuitive
%   fact, but it's not worth it to try to make \cs\foreign and \cs\word smart
%   enough to see each other inside themselves.}
%
% \DescribeMacro{\term}
%    \cs\term\marg{technical term} typesets a techincal term in a different
%    font.  You might want to use this where a techincal term is first used, or
%    defined.  One could enhance this macro and \cs\defn to help build an
%    automatic glossary
%
%    \begin{bothexample}
%      This sort of thing is called a \term{blibnil}.
%    \end{bothexample}
%
% \DescribeMacro{\defn}
%   \cs\defn\marg{definition} typesets a definition, perhaps of a technical
%   term.  One could enhance this macro and \cs\defn to help build an automatic
%   glossary
%
%   \begin{bothexample}
%     We may describe a \term{blibnil} as \defn{a slibnil with
%     three arms}.
%   \end{bothexample}
%
% \section{Titles}
%
% \DescribeMacro\book
%   \cs\book\marg{book title} typesets a book title.
%
%   \begin{bothexample}
%     Some people find \book{Moby-Dick} dull, but I thought
%     it was exciting.
%   \end{bothexample}
%
% \DescribeMacro\journal
%   \cs\journal\marg{journal title} typesets a journal title.
%
%   \begin{bothexample}
%     I liked it so much I started a scholarly journal called
%     \journal{The Melville Times} with the inheritance from
%     my grandmother.
%   \end{bothexample}
%
% \DescribeMacro\music
%   \cs\music\marg{music title} typesets a music title.  
%
%   \begin{bothexample}
%     My journal didn't do very well; I moped around my office
%     and listened to Schubert's \music{Winterreise}.
%   \end{bothexample}
%
% \DescribeMacro\article
%   \cs\article\marg{article title} typesets a article title.  
%
%   \begin{bothexample}
%     Then one day I received an article, \article{Pip and
%     the Milk of Human Kindness}, by express mail from Wales.
%   \end{bothexample}
%
% \DescribeMacro\poemtitle
%   \cs\poemtitle\marg{poem title} typesets a poem title.  
%
%   \begin{bothexample}
%     I then wrote my famous poem \poemtitle{Jump for Joy like
%     the Butterflies of Troy} in five minutes.
%   \end{bothexample}
%
%   Sometimes longer poems are distinguished from shorter ones in type when
%   they have been published separately as a book [FIX give reference].  This
%   package defines a macro \cs\longpoem in the configuration file in the
%   following way:
%   \begin{codeexample}
%     \newlet\longpoem\textitswitch
%   \end{codeexample}
%
% \DescribeMacro\play
%   \cs\play\marg{play title} typesets a play title.  
%
%   \begin{bothexample}
%     To celebrate the popularity of the article, I took the
%     author to the theater to see the acclaimed play
%     \play{Grave in Waterloo}, starring Vincent Price.
%   \end{bothexample}
%  
% \DescribeMacro\craft
%   \cs\craft\marg{craft title} typesets a title of a craft or ship.
%
%   \begin{bothexample}
%     With tears in my heart, I put the author on the \craft{HMS Shangrila}
%     bound for Wales.
%   \end{bothexample}
%  
% \DescribeMacro\species
%   \cs\species\marg{[ genus ] species [ subspecies ]} typesets the Latin
%   generic and/or specific names for an organism.
%
%   \begin{lesson}
%     \manual specifies italic type.  Genus names should be capitalized, and
%     may be abbreviated on subsequent appearances with the initial letter.
%     Following designations should be in roman.  E.g., ``var.''\ for
%     ``variant'' following species name and ``sp.''\ for ``species'' following
%     genus name, meaning ``any species in the genus.''  \attrib{7.102--4}
%
%     Higher groupings should be in capitalized roman.  English derivatives of
%     scientific names, e.g., amoeba, are lowercased.  \attrib{7.105--6}
%                                                      
%     \todo{\package{abbrevs}
%     category for genus/species and/or datemark for suffixes}
%   \end{lesson}
%
% \begin{warning}
%   Right now there is a small discrepancy between the behavior of
%   \cs\textitswitch and \cs\Wrapquotes regarding what happens when followed by
%   a command sequence such as \cs\footnote.  I hope to make these things
%   completely parallel one day, but for now, realize that after using a
%   titling macro that uses \cs\Wrapquotes, you must use |{}| before any
%   following command sequence that you want to immediately follow the title
%   with no intervening space.  The only case I can think of is \cs\footnote.
%   If you forget the |{}|, you will have an extra space after the title and before the
%   footnotemark.  The following example illustrates this behavior and
%   contrasts it with \cs\textitswitch:
% \begin{bothexample}
%   \newabbrev\foo{Foo}
%   
%   \book{Foo}\foo
%   
%   \book{Foo} \foo
%   
%   \book{Foo}\footnote{footie}
%
%   \book{Foo}{}\footnote{footie}
%
%   \poemtitle{Foo}\foo
%
%   \poemtitle{Foo} \foo
%
%   \poemtitle{Foo}\footnote{footie}
%
%   \poemtitle{Foo}{}\footnote{footie}
%
%   \poemtitle{Foo}.\footnote{footie}
% \end{bothexample}
% \end{warning}
%
% \section{Programmer's interface}
%
%  \DescribeMacro\Wrapquotes
%  \DescribeMacro\WrapquotesNS
%  \DescribeMacro\WrapquotesIS
%  \DescribeMacro\WrapquotesNN
%  \DescribeMacro\WrapquotesIN
%  \DescribeMacro\WrapquotesSN
%  \DescribeMacro\WrapquotesDN
%  \DescribeMacro\WrapquotesSK
%    \cs\Wrapquotes\marg{text} wraps \meta{text} in quotes.  Single quotes are
%    used initially if the \option{singlequotes} option is given to the
%    package, and double quotes if no option or the \option{doublequotes}
%    option is given to the package.  
%     
%    When quotation marks inserted by \cs\Wrapquotes and friends are doubled up
%    (this occurs sometimes when nesting them), a thinspace (|\,|) is inserted
%    between the abutted quotes.
%
%    \cs\Wrapquotes will be \cs\let to one of the six macros
%    \cs\Wrapquotes\meta{XY}. 
%
%    In the two-letter suffix \meta{XY}, first letter |N| means ``normal'' and
%    |I| means ``inverse.''  These are macros that switch between single and
%    double quotes when they nest: an inverse wrapquotes wraps with single
%    quotes when a normal wrapquotes would wrap with double quotes, and vice
%    versa.  First letter |S| for ``single'' and |D| for ``double'' are for
%    macros that always wrap with single or double quotes.  Spacing and
%    punctuation following the closing quotes are handled intelligently by
%    macros with second letter |S|, which means means suck a following period
%    or comma into the closing quote, that is, if what follows is a comma or
%    period, it is pulled inside the quotes (following American practice).
%    Second letter |N| means ``nosuck,'' that is, don't suck.  Second letter
%    |K| means ``kill'': the same as |N| but suppress the effect of any
%    punctuation in the quoted argument on spacing that follows the closing
%    quotes (i.e., execute \cs\@, which sets the spacefactor to~1000).  This is
%    only useful in certain technical writing where punctuation in the quoted
%    argument should not be considered puncutation of the containing sentence.
%
%    A space is inserted after the closing quotes unless what follows is in the
%    set |;?:!-)]'\textquoteright{|, in which case no space is inserted.^^A for Emacs: }
%  FIX: that would be |\nospacelist|
%
%  \DescribeMacro\IfQuestionOrExclamation
%    \cs\IfQuestionOrExclamation\marg{text}\marg{true}\marg{false} 
%    executes \meta{true} clause iff \meta{text} ends with a question mark or
%    an exclamation point; executes \meta{false} clause otherwise.
%
%   \subsection{Limitations of Wrapquotes and friends}
%    ^^A FIX: hmm, can't use \cs or \protect\cs
%      \subsubsection{Nesting}
%     \begin{warning}
%       For proper nesting of \cs\Wrapquotes and friends, user commands must be
%       \cs\let to \cs\Wrapquotes or one of the six \cs\Wrapquotes\meta{XY}
%       commands instead of using a \cs\def-like defining command.  It's OK to
%       \cs\let % a user macro to something like \cs\Wrapquotes which itself
%       has been % \cs\let to one of the six \cs\Wrapquotes\meta{XY} macros.
%
%       The user command which is \cs\let to one of the \cs\Wrapquotes commands
%       must furthermore appear in the source.  That is, it must not appear as
%       the result of an expansion.  Among other things, this means that
%       nesting won't work properly if you put \cs\Wrapquotes into an abbrev
%       (see the \package{abbrevs} package in the \Frankenstein bundle).
%
%       For applications where nesting will not occur, there should be nothing
%       to worry about.
%    \end{warning}
%
%      \subsubsection{Italic corrections}
%
%      \begin{warning}
%        The question of when to insert an italic correction is not nearly as
%        simple as it might seem.  I cannot figure good rules which cover all
%        cases, and I do not trust the behavior of the kernel's macros as a
%        guide.  So I can not tell you whether this package handles italic
%        corrections properly.  If you discover behavior which you think is
%        wrong, please let me know with an example and an argument.
%      \end{warning}
%
%      \subsubsection{A slight bug}
%      \begin{warning}
%        Right now there is a small bug in cases where closing quotes fall at
%        the end of italic text, such as
%        \begin{bothexample}
%          \normalfont
%          \book{My love of \poemtitle{Daffodils}}, by H.~Moneysworth.
%        \end{bothexample}
%        These cases loose because the closing quotation marks and any
%        sucked-in punctuation are going to be in roman, not italic, or italic,
%        not roman.  Only the more obsessive will notice this flaw.  I'm sure I
%        will come up with a way to handle this for a future version of this
%        package.
%      \end{warning}
%
% \StopEventually{}
%
% \part{Implementation}
%
% \section{Version control}
%
%  \begin{macro}{\fileinfo}
%  \begin{macro}{\DoXUsepackagE}
%  \begin{macro}{\HaveECitationS}
%  \begin{macro}{\fileversion}
%  \begin{macro}{\filedate}
%  \begin{macro}{\docdate}
%  \begin{macro}{\PPOptArg}
% These definitions must be the first ones in the file.
%    \begin{macrocode}
\def\fileinfo{title macros (Frankenstein's references)}
\def\DoXPackageS {}
\def\initelyHavECitationS {}
\def\fileversion{v1.2}
\def\filedate{2001/08/31}
\def\docdate{2001/08/31}
\edef\PPOptArg {%
  \filedate\space \fileversion\space \fileinfo
}
%    \end{macrocode}
%  \end{macro}
%  \end{macro}
%  \end{macro}
%  \end{macro}
%  \end{macro}
%  \end{macro}
%  \end{macro}
%
% If we're loading this file from a \cs\ProcessDTXFile command (see the
% \package{compsci} package), then \cs\JusTLoaDInformatioN will be defined;
% othewise we assume it is not (that's why the FunkY NamE).
% 
% If we're loading from \cs\ProcessDTXFile, we want to load the packages listed
% in \cs\DoXPackageS (needed to typeset the documentation for this file) and
% then bail out.  Otherwise, we're using this file in a normal way as a
% package, so do nothing.  \cs\DoXPackageS, if there are any, are declared in
% the \ext{dtx} file, and, if you're reading the typeset documentation of this
% package, would appear just above.  (It's OK to call \cs\usepackage with an
% empty argument or \cs\relax, by the way.)
%    \begin{macrocode}
\makeatletter% A special comment to help create bst files.  Don't change!
\@ifundefined{JusTLoaDInformatioN} {%
  }{% ELSE (we know the compsci package is already loaded, too)
  \UndefineCS\JusTLoaDInformatioN
  \SaveDoXVarS
  \eExpand\csname DoXPackageS\endcsname\In {%use \csname in case it's undefined
    \usepackage{#1}%
  }%
  \RestoreDoXVarS
  \makeatother
  \endinput
}% A special comment to help create bst files.  Don't change!
%    \end{macrocode}
%
% Now we check for \LaTeX2e and declare the LaTeX package.
%    \begin{macrocode}
\NeedsTeXFormat{LaTeX2e}
\ProvidesPackage{titles}[\PPOptArg]
%    \end{macrocode}^^A special comment to help create bst files.  Don't change!
%
% ^^A NOTE: We have to compensate for the above backslashes, which are not
% ^^A       actually in the .dtx file the author works on, by adding to the
% ^^A       CheckSum.
%%
% \AddToCheckSum{17}^^A `dtx-update-checksum' automatically handles this.
% \AddToCheckSum{7}^^A The half a macrocode env. at the top is missed, however...
% \AddToCheckSum{10}^^A ... and so are the 5 \defs from the .dtx file 
%                   ^^A     that precede it.
% \IfCitations {%
%   \AddToCheckSum{2}^^A When \initelyHavECitationS is defined in
% }                  ^^A the .dtx file, we need 2 more in the CheckSum.
%
%
% \section{Requirements}
%
%    \begin{macrocode}
\RequirePackage{moredefs,slemph}
%    \end{macrocode}
%
% \section{Options}
% 
% \begin{macro}{\ti@domelater}
%    \begin{macrocode}
\ReserveCS\ti@domelater
\DeclareOption{british} {%
  \def\ti@domelater {%
    \let\Wrapquotes\WrapquotesNN
    \@doublequotes@false
  }
}
\DeclareOption{american} {%
  \def\ti@domelater {%
    \let\Wrapquotes\WrapquotesNS
    \@doublequotes@true
  }
}
\ExecuteOptions{american}
\ProcessOptions
%    \end{macrocode}
% \end{macro}
%
% \section{Wrapquotes}
%
% Here we go!  This is not a picnic, so leave your jelly jar home.
%
%   \subsection{Titles that are Questions or Exclamations}
%
% \begin{macro}{\IfQuestionOrExclamation}
% \begin{macro}{\ti@checkfor@q}
% \begin{macro}{\ti@checkfor@e}
% \begin{macro}{\ti@prev}
% \begin{macro}{\ti@prev@prev}
% \begin{macro}{\@ti@sw@true}
% \begin{macro}{\@ti@sw@false}
% \begin{macro}{\if@ti@sw@}
%   \mbox{}
%    \begin{macrocode}
\newcommand\IfQuestionOrExclamation [1] {%
  \@tempswafalse
  \ti@checkfor@q #1?\@nil
  \ti@checkfor@e #1!\@nil
  \if@tempswa
    \expandafter\@firstoftwo
  \else
    \expandafter\@secondoftwo
  \fi
}
%    \end{macrocode}
% The large majority of titles will not contain a question mark or exclamation
% point.  The large majority of those that do will have a single mark or point
% at the end.  We could (I think) use a simpler check that processed all titles
% by looping through to examine the end, but a slightly more complicated check
% will handle the majority of cases very quickly (and at a constant speed,
% rather than proportional to title length) and not greatly slow down
% processing the remaining two unusual cases.  We divide our argument (with an
% extra question mark tacked onto the end) into what's before the first
% question mark and what's after it.  Then we examine what's after it and
% interpret the results thus:
% \begin{description}
% \item [empty] no question mark in title
% \item [question mark] title ends with question mark (and there are no other
%       question marks)
% \item [text ending with one question mark] a question mark occurs in the
%       title, but not at the end
% \item [text ending with two question marks] title ends with a question mark
%       (and there is a previous question mark)
% \end{description}
% We set switch |a| to true if the title ends with a question mark.
%    \begin{macrocode}
\newboolean{@ti@sw@}
\ReserveCS\ti@prev
\ReserveCS\ti@prev@prev
\NewName{ti@checkfor@q} {#1?#2\@nil} {%
  \def\sc@t@a{#2}%  
  \def\sc@t@b{?}%
  \ifx\sc@t@a\ShortEmpty
  \else
    \ifx\sc@t@a\sc@t@b
      \@tempswatrue
    \else
%    \end{macrocode}
% We use a loop to whittle down |#2| until \cname{ti@prev} contains the last
% character and \cname{ti@prev@prev} contains the second-to-last.  We know that
% \cname{ti@prev} is going to be a question mark.  Iff \cname{ti@prev@prev} is a
% question mark, we are in the final case above.
%    \begin{macrocode}
      \let\ti@prev\sc@t@a
      \let\ti@prev@prev\sc@t@a
      \@ti@sw@true
      \@whilesw \if@ti@sw@ \fi {%
        \ifx\sc@t@a\ShortEmpty
          \@ti@sw@false
        \else
          \let\ti@prev@prev\ti@prev
          \let\ti@prev\sc@t@a
          \edef\sc@t@a{\E@cdr\sc@t@a\@nil}%
        \fi
      }%
      \edef\ti@prev@prev{\E@car\ti@prev@prev\@nil}%
      \ifx\ti@prev@prev\sc@t@b
        \@tempswatrue
      \fi
    \fi
  \fi  
}
%    \end{macrocode}
% Exact same logic applies to exclamation points.
%    \begin{macrocode}
\NewName{ti@checkfor@e} {#1!#2\@nil} {%
  \def\sc@t@a{#2}%  
  \def\sc@t@b{!}%
  \ifx\sc@t@a\ShortEmpty
  \else
    \ifx\sc@t@a\sc@t@b
      \@tempswatrue
    \else
      \let\ti@prev\sc@t@a
      \let\ti@prev@prev\sc@t@a
      \@ti@sw@true
      \@whilesw \if@ti@sw@ \fi {%
        \ifx\sc@t@a\ShortEmpty
          \@ti@sw@false
        \else
          \let\ti@prev@prev\ti@prev
          \let\ti@prev\sc@t@a
          \edef\sc@t@a{\E@cdr\sc@t@a\@nil}%
        \fi
      }%
      \edef\ti@prev@prev{\E@car\ti@prev@prev\@nil}%
      \ifx\ti@prev@prev\sc@t@b
        \@tempswatrue
      \fi
    \fi
  \fi  
}
%    \end{macrocode}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
%
%   \subsection{Highlevel macros}
%
% \begin{macro}{\ti@wrapquotes@suck}
% \begin{macro}{\ti@wrapquotes@nosuck}
%   These two are the top-level internal macros, and they are pretty sane.  One
%   sucks in a following period or comma, the other does not.
%   \cname{ti@wrapquotes@suck} does not suck, however, when the title ends in a
%   question or exclamation point.
%
%  The group here is necessary to scope the \cname{@doublequotes@} boolean.
%    \begin{macrocode}
\newcommand*\ti@wrapquotes@suck [1] {%
%  \DTypeout{top of wrapquotes@suck}%
  \IfQuestionOrExclamation {#1} {%
      \ti@wrapquotes@nosuck{#1}%
    }{% ELSE
%    \DTypeout{top of wrapquotes@suck ELSE}%
    \begingroup
      \if@doublequotes@
%        \DTypeout{double true in suck}%
        \@doublequotes@false
        \def\sc@t@a {\ti@open@double #1\ti@close@double@suck}%
      \else
%        \DTypeout{double false in suck}%
        \@doublequotes@true
        \def\sc@t@a {\ti@open@single #1\ti@close@single@suck}%
      \fi
      \sc@t@a
    \endgroup
    }%
}
\newcommand*\ti@wrapquotes@nosuck [1] {%
  \begingroup
    \if@doublequotes@
%      \DTypeout{double true in nosuck}%
      \@doublequotes@false
      \def\sc@t@a {\ti@open@double #1\ti@close@double@nosuck}%
    \else
%      \DTypeout{double false in nosuck}%
      \@doublequotes@true
      \def\sc@t@a {\ti@open@single #1\ti@close@single@nosuck}%
    \fi
    \sc@t@a
  \endgroup
}
%    \end{macrocode}
% \end{macro}
% \end{macro}
%
% \begin{macro}{\WrapquotesNS}
% \begin{macro}{\WrapquotesIS}
% \begin{macro}{\WrapquotesNN}
% \begin{macro}{\WrapquotesIN}
% \begin{macro}{\WrapquotesSN}
% \begin{macro}{\WrapquotesDN}
% \begin{macro}{\WrapquotesSK}
% \begin{macro}{\Wrapquotes}
% \begin{macro}{\if@doublequotes@}
% \begin{macro}{\@doublequotes@true}
% \begin{macro}{\@doublequotes@false}
%   Now we can define the secondary programmers' macros.
%
%   We simply reserve \cs\Wrapquotes here, and assign it in the user options
%   section above.
%    \begin{macrocode}
\newboolean{@doublequotes@}
\newcommand*\WrapquotesNS {%
%  \DTypeout{starting wrapquotes NS}%
  \ti@wrapquotes@suck
}
\newcommand*\WrapquotesIS {%
%  \DTypeout{starting wrapquotes IS}%
  \ToggleBoolean{@doublequotes@}%
  \ti@wrapquotes@suck
}
\newcommand*\WrapquotesNN {%
%  \DTypeout{starting wrapquotes NN}%
  \ti@wrapquotes@nosuck
}
\newcommand*\WrapquotesIN {%
%  \DTypeout{starting wrapquotes IN}%
  \ToggleBoolean{@doublequotes@}%
  \ti@wrapquotes@nosuck
}
\newcommand*\WrapquotesSN [1] {%
%  \DTypeout{starting wrapquotes SN}%
  \begingroup
    \ti@open@single #1\ti@close@single@nosuck
  \endgroup
}
\newcommand*\WrapquotesDN [1] {%
%  \DTypeout{starting wrapquotes DN}%
  \begingroup
    \ti@open@double #1\ti@close@double@nosuck
  \endgroup
}
\newcommand*\WrapquotesSK [1] {% FIX: test
%  \DTypeout{starting wrapquotes SK}%
  \begingroup
    \ti@open@single #1\ti@close@single@nosuck\@%
  \endgroup
}
\ReserveCS\Wrapquotes
\ti@domelater
%    \end{macrocode}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
%
%   \subsection{Opening quotes}
%
% \begin{macro}{\ti@open@double}
% \begin{macro}{\ti@open@single}
% \begin{macro}{\ti@openquote}
%   We start by putting an opening mark in scratch |f| with a global
%   definition.  
%
%   I can't remember why it's global.  In the macros that close quotes, we want
%   to keep that information around past a group end because we're using
%   \cs\aftergroup, but that doesn't seem to apply for opening them.  Best not
%   to change what's not broke, however.
%    \begin{macrocode}
\newcommand\ti@open@double {%
  \gdef\sc@t@f {\textquotedblleft}%
  \ti@openquote
}
\newcommand\ti@open@single {%
  \gdef\sc@t@f {\textquoteleft}%
  \ti@openquote
}
%    \end{macrocode}
% Then we look ahead with scratch |a|.  We are looking ahead at the first
% character of the contents of the \cs\Wrapquotes.
%    \begin{macrocode}
\newcommand\ti@openquote {% 
  \futurelet\sc@t@a\ti@@openquote
}
%    \end{macrocode}
% Insert the opening mark.  Then, if we are about to open another quote, insert
% the space appropriate to separate contiguous quotation marks.
%    \begin{macrocode}
\newcommand\ti@@openquote {%
  \sc@t@f
  \ifx\sc@t@a\WrapquotesNS
%    \DTypeout{Quotation marks are doubled up (next is NS); inserting padding.}%
    \,%
  \else \ifx\sc@t@a\WrapquotesNN
%    \DTypeout{Quotation marks are doubled up (next is NN); inserting padding.}%
    \,%
  \else \ifx\sc@t@a\WrapquotesIN
%    \DTypeout{Quotation marks are doubled up (next is IN); inserting padding.}%
    \,%
  \else \ifx\sc@t@a\WrapquotesIS
%    \DTypeout{Quotation marks are doubled up (next is IS); inserting padding.}%
    \,%
  \else \ifx\sc@t@a\WrapquotesSN
%    \DTypeout{Quotation marks are doubled up (next is SN); inserting padding.}%
    \,%
  \else \ifx\sc@t@a\WrapquotesDN
%    \DTypeout{Quotation marks are doubled up (next is DN); inserting padding.}%
    \,%
  \else \ifx\sc@t@a\WrapquotesSK
%    \DTypeout{Quotation marks are doubled up (next is SK); inserting padding.}%
    \,%
  \else
  \fi \fi \fi \fi \fi \fi \fi
}
%    \end{macrocode}
% \end{macro}
% \end{macro}
% \end{macro}
%
%   \subsection{Closing macros that don't suck}
%
% This case that doesn't suck is easier, so we warm up with it.
%
% \begin{macro}{\ti@close@single@nosuck}
% \begin{macro}{\ti@close@double@nosuck}
% \begin{macro}{\ti@close@single@@nosuck}
% \begin{macro}{\ti@close@double@@nosuck}
%   \mbox{}   
%    \begin{macrocode}
\newcommand*\ti@close@single@nosuck {%
  \aftergroup\ti@close@single@@nosuck
}
\newcommand*\ti@close@double@nosuck {%
  \aftergroup\ti@close@double@@nosuck
}
\newcommand*\ti@close@single@@nosuck {%
  \gdef\sc@t@f {\textquoteright}%
  \ti@close@quote@nosuck
}
\newcommand*\ti@close@double@@nosuck {%
  \gdef\sc@t@f {\textquotedblright}%
  \ti@close@quote@nosuck
}
%    \end{macrocode}
%  \end{macro}
%  \end{macro}
%  \end{macro}
%  \end{macro}
%
% \begin{macro}{\ti@close@quote@nosuck}
% \begin{macro}{\if@look@nosuck@}
% \begin{macro}{\@look@nosuck@true}
% \begin{macro}{\@look@nosuck@false}
%    \todo{Document this flag.  It's a hack, we must set it before each call
%    to \cname{ti@q@ifnextcharin} I think.  What it stands for is something
%    like the presence of the tokens \cname{ti@close@single@nosuck} and
%    \cname{ti@close@double@nosuck} in the list of chars to look for, but since
%    they aren't really chars they can't go in the list, so instead we set the
%    flag.  Somewhat cleaner would be putting a flag char in the list, but I
%    can't think of what char I could safely use.}
%    \begin{macrocode}
\newboolean{@look@nosuck@}
\@look@nosuck@false
\newcommand\ti@close@quote@nosuck {%
%  \DTypeout{Starting ti@close@quote@nosuck}%
  \@look@nosuck@true
%    \end{macrocode}
%% FIX
% Aha, but here is a good reason to leave in |.,| in our substitute for \cs\nospacelist.
%    \begin{macrocode}
  \expandafter \ti@q@ifnextcharin \expandafter {\nospacelist} {%
%      \DTypeout{Found a nosuck no-spacer.  C=[\meaning\sc@t@c] F=[\meaning\sc@t@f]}%
      \sc@t@f
    }{% ELSE
%      \DTypeout{Found a nosuck spacer. C=[\meaning\sc@t@c] F=[\meaning\sc@t@f]}%
      \sc@t@f\space
  }%
}
%    \end{macrocode}
%  \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
%
%   \subsection{Closing macros that suck}
%
% \begin{macro}{\ti@close@double@suck}
% \begin{macro}{\ti@close@single@suck}
%   We need to look ahead beyond the \cs\endgroup that ends
%   \cname{Wrapquotes??}.  The lookahead mechanism that gets invoked in scratch
%   |a| below could handle looking past the \cs\endgroup, but I think it is
%   more efficient to skip it by using \cs\aftergroup.
%    \begin{macrocode}
\newcommand\ti@close@double@suck {%
  \aftergroup\ti@close@double@@suck
}
\newcommand\ti@close@single@suck {%
  \aftergroup\ti@close@single@@suck
}
%    \end{macrocode}
% \end{macro}
% \end{macro}
%
% \begin{macro}{\ti@close@double@@suck}
% \begin{macro}{\ti@close@single@@suck}
%   This part isn't so bad yet.  To close the quotes, we again start with the
%   closing mark in scratch |f|, with a global definition.
%    \begin{macrocode}
\newcommand\ti@close@double@@suck {%
  \gdef\sc@t@f {\textquotedblright}%
  \ti@close@quote@suck
}
\newcommand\ti@close@single@@suck {%
  \gdef\sc@t@f {\textquoteright}%
  \ti@close@quote@suck
}
%    \end{macrocode}
% \end{macro}
% \end{macro}
%
% \begin{macro}{\nospacelist}
%   Put these in the order of their frequency.  Anything in \cs\nocorrlist
%   should also be in here, most likely.  I'm putting in \cname{@xobeysp}
%   because it's in the \package{xspace} package, but I can't tell you when it
%   would come up.
%    \begin{macrocode}
\requirecommand\nospacelist {%
  ,.':;?-/\slash~!)]\bgroup\egroup\@sptoken\ \space\/\@xobeysp
}
%    \end{macrocode}
% \end{macro}
%
% \begin{macro}{\ti@close@quote@suck}
%   Then we use \cname{ti@q@ifnextcharin} to look as far ahead as necessary for
%   a significant character.  The latest significant character found is
%   available in scratch |c|.  The work of handling all the cases of what we
%   might find while looking ahead is divided up between
%   \cname{ti@close@quote@suck} and \cname{ti@q@ifnextcharin}.
%   \cname{ti@close@quote@suck} handles the last step in the process, and
%   \cname{ti@q@ifnextcharin} handles all the steps up to the last.
%
%   Here is what \cname{ti@close@quote@suck} does, in English.  If we find a
%   comma or period, we put it inside the closing quote, and gobble the one we
%   found.  That is, we print out scratch |c|, then scratch |f|, then gobble a
%   character.  If we find something in the set given in \cs\nospacelist, do
%   not leave a space after the closing mark.  That is, just print out scratch
%   |f|. If we find something else, we leave a space after the closing mark.
%   That is, print scratch |f| and a space.
%    \begin{macrocode}
\newcommand\ti@close@quote@suck {%
%  \DTypeout{Starting ti@close@quote@suck}%
  \@look@nosuck@false
  \ti@q@ifnextcharin {.,} {%
%      \DTypeout{Found a comma or period. C=[\meaning\sc@t@c] F=[\meaning\sc@t@f]}%
      \sc@t@c\sc@t@f\DGobbleM % This gobbles the original punctuation.
    }{% ELSE
%    \DTypeout {Before second ti@qifnextcharin.  C=[\meaning\sc@t@c] F=[\meaning\sc@t@f]}%
    \@look@nosuck@true
%    \end{macrocode}
% \begin{todoenv}
%   Using \cs\nospacelist is inefficient here, since some of the cases,
%^^A for Emacs {
%   namely \verb*|,.\@sptoken}|, are never going to be there and shouldn't be
%   checked for, since they are passed over by \cname{ti@q@ifnextcharin} before
%   the list is compared.  But it would be good to have this parallelism between
%   \package{abbrevs} and \package{titles}.
% \end{todoenv}
%    \begin{macrocode}
    \expandafter \ti@q@ifnextcharin \expandafter {\nospacelist} {%
%        \DTypeout{Found a suck no-spacer.  C=[\meaning\sc@t@c] F=[\meaning\sc@t@f]}%
        \sc@t@f
      }{% ELSE
%      \DTypeout{Found a suck spacer. C=[\meaning\sc@t@c] F=[\meaning\sc@t@f]}%
      \sc@t@f\space
    }%
  }%
}
%    \end{macrocode}
%  \end{macro}
%
%   \subsection{Looking ahead}
%
%   Now things are getting fun.
%
% \begin{macro}{\ti@q@ifnextcharin}
% \begin{macro}{\ti@q@check}
% \begin{macro}{\ti@q@ifnch}
% \begin{macro}{\ti@q@@ifnch}
%   These macros are modeled after the definition of \cname{@ifnextchar} which
%   skips spaces.  While looking ahead for the next significant character,
%   these macros skip spaces, \cs\egroup, \cs\endgroup, \cname{check@icr},
%   \cname{ti@close@double} and \cname{ti@close@single} while doing the right
%   thing after each.
%
%   The first argument should be a list of tokens.  If the next significant
%   char is in the list, then the |true| clause is executed, otherwise the
%   |false| clause is executed.  The next significant char is left in scratch
%   |c| so it can be accessed by the clauses.
%
%   The three arguments to \cname{ti@q@ifnextcharin} are saved in global
%   variables because while looking ahead we must continue past the ends of
%   groups.
%
%% FIX
% Not sure I need gdef for scratch |e|.
%    \begin{macrocode}
\newcommand\ti@q@ifnextcharin [3] {% args: charlist true false
%  \sc@toks@a{#1}%
%  \DTypeout{charlist unexpanded is =[\the\sc@toks@a]}%
  \gdef\sc@t@e {#1}%
  \gdef\sc@t@a {#2}%
  \gdef\sc@t@b {#3}%
  \ti@q@check
}
%    \end{macrocode}
% Having saved the arguments, we look ahead with scratch |c|.  This step is not
% in the macro above so that we can jump back to \cname{ti@q@check} whenever we
% want to look ahead another character.
%    \begin{macrocode}
\newcommand\ti@q@check {%
  \futurelet\sc@t@c\ti@q@ifnch
}
%    \end{macrocode}
% Scratch |c| contains the current char.  Scratch |d| is the action to take at
% the end of this macro.  We attempt to order these possibilities to make
% \cs\Wrapquotes most efficient, though it is a guess which items will be
% encountered most frequently.
% 
% The actions taken for each of the possibilities are the following:
% \begin{description}
%   \item[\cs\ifvmode]  Assume that the \cs\Wrapquotes was the argument of a
%     \cs\TextFontCommand from certain \LaTeX{} kernels.  Gobble three more tokens
%     expected to follow the \cs\ifvmode, execute them, and continue on to look
%     ahead another character.  See documentation of
%     \cname{ti@q@handle@ifvmode} for more details.
%   \item[\cname{check@icr}] This means the \cs\Wrapquotes was the 
%     argument of a \cs\TextFontCommand.  Gobble the \cname{check@icr} 
%     and look ahead another character after we exit the
%     group that the \cs\TextFontCommand has given us.
%   \item[\cs\endgroup and {\normalfont\ttfamily \}}] Pass right by an \cs\endgroup or
%     |}| and look ahead another char.
%   \item[\cname{@sptoken} (a non-explicit space)] Handle a non-explicit space by calling
%     \cname{ti@q@handle@space}, which gobbles the space and looks ahead
%     another char.  When the user or a macro has followed the titles with an
%     \emph{explicit} space such as a tie, or the
%%    \verb*|\ |
%     or \cs\space macros, we do nothing and let this be caught by the
%     comparison to the tokens in the argument of \cname{ti@q@ifnextcharin}.
%   \item[\cname{ti@close@double@suck},
%         \cname{ti@close@single@suck}]
%   \item[\cname{ti@close@double@nosuck}, and
%         \cname{ti@close@single@nosuck}]  We are in a nested
%         \cs\Wrapquotes.  Call \cname{ti@q@handle@single/double@suck/nosuck} 
%         as appropriate, which 
%         gobbles the closequotes token, adds properly-padded closing quotes to
%         scratch |f|, and then goes on to look ahead another character.
% \end{description}
% The lookahead process stops when it finds something not on this list.  Then
% it compares what it found to the list of characters given to
% \cname{ti@q@ifnextcharin} and executes the |true| or |false| clause as
% appropriate.
%
% First we have to handle the case of finding an \cs\ifvmode.  We can't bundle
% this test in with the tests for other tokens, so it gets its own macro,
% \cname{ti@q@handle@ifvmode}, which see for details.  The remaining cases are
% handled in \cname{ti@q@@ifnch}.  
%    \begin{macrocode}
\newcommand\ti@q@ifnch {%
%  \DTypeout{The lookahead in ti@q@ifnch: [\meaning\sc@t@c]}%
  \ifx\sc@t@c\ifvmode
    \let\sc@t@d\ti@q@handle@ifvmode
  \else
    \let\sc@t@d\ti@q@@ifnch
  \fi
  \sc@t@d
}
\newcommand\ti@q@@ifnch {%
%  \DTypeout{entering ti@q@@ifnch}%
%  \expandafter\sc@toks@a\expandafter{\sc@t@c}%
%  \DTypeout{ti@q@@ifnch: C expanded once is =[\the\sc@toks@a]}%
  \ifx\sc@t@c\check@icr
%    \DTypeout{Handling check@icr}%
    \defcommand\sc@t@d [1] {%
%      \DTypeout{check@icr handler: gobbling [\meaning ##1]}%
      ##1\aftergroup\ti@q@check
    }%
  \else \ifx\sc@t@c\endgroup 
%    \DTypeout{Handling endgroup}%
    \def\sc@t@d {\aftergroup\ti@q@check}%
  \else \ifx\sc@t@c\@sptoken
%    \DTypeout{Handling space}%
    \let\sc@t@d\ti@q@handle@space
  \else \ifx\sc@t@c\egroup
%    \DTypeout{Handling egroup}%
    \def\sc@t@d {\aftergroup\ti@q@check}%
  \else \ifx\sc@t@c\ti@close@double@suck
%    \DTypeout{Handling ti@close@double@suck}%
    \let\sc@t@d\ti@q@handle@double@suck
  \else \ifx\sc@t@c\ti@close@single@suck
%    \DTypeout{Handling ti@close@single@suck}%
    \let\sc@t@d\ti@q@handle@single@suck
  \else \ifx\sc@t@c\ti@close@double@nosuck
%    \DTypeout{Handling ti@close@double@nosuck}%
    \let\sc@t@d\ti@q@handle@double@nosuck
  \else \ifx\sc@t@c\ti@close@single@nosuck
%    \DTypeout{Handling ti@close@single@nosuck}%
    \let\sc@t@d\ti@q@handle@single@nosuck
  \else
%    \end{macrocode}
% We've handled all the lookahead cases, so now we are left with the simple
% comparison of the next char with the charlist.
%    \begin{macrocode}
              \@tempswafalse
              \expandafter   \@tfor
                \expandafter \sc@t@g
                \expandafter :%
                \expandafter =%
                             \sc@t@e
                             \do {%
                  \expandafter\ifx\sc@t@g\sc@t@c
%                  \DTypeout{We have a match of [\meaning\sc@t@c]
%                           with [\expandafter\meaning\sc@t@g]}%
                  \@tempswatrue
                  \@break@tfor
                \else
%                  \DTypeout{We have NO match between [\meaning\sc@t@c]
%                           with [\meaning\sc@t@g]}%
                \fi
              }%
              \if@tempswa
%                \DTypeout{Choosing true clause [\meaning\sc@t@a]}%
                \let\sc@t@d\sc@t@a % the ``true'' clause
              \else
%                \DTypeout{Choosing false clause [\meaning\sc@t@b]}%
                \let\sc@t@d\sc@t@b % the ``false'' clause
              \fi
  \fi \fi \fi \fi \fi \fi \fi \fi
%  \DTypeout{About to fall out of ti@q@@ifnch and do this [\meaning\sc@t@d]}%
  \sc@t@d
}
%    \end{macrocode}
%  \end{macro}
%  \end{macro}
%  \end{macro}
%  \end{macro}
%
% \begin{macro}{\ti@q@handle@ifvmode}
%   This is in its own macro for clarity and to avoid problems with skipping
%   over clauses.
%
%   \cname{ti@q@ifnch} has to take two different kinds of \LaTeX{} kernel into
%   account.  The 1996/12/01 and 1997/06/01 kernels used a different definition
%   of \cs\DeclareTextFontCommand:
%    \begin{macrocode}
%\def \DeclareTextFontCommand #1#2{%
%  \DeclareRobustCommand#1[1] {%
%    \ifmmode
%      \nfss@text{#2##1}%
%    \else
%      \leavevmode
%      {\text@command{##1}%
%       #2\check@icl ##1\ifvmode\else\check@icr\fi
%       \expandafter}%
%    \fi
%  }%
%}
%    \end{macrocode}
%   All other kernels leave out the check for vertical mode (kernels from
%   1997/12/01 include it when necessary inside \cname{check@icr}).  The macro
%   \cname{ti@q@ifnch}, which will be called immediately before this point of
%   difference, handles both cases by looking for both \cs\ifvmode and
%   \cname{check@icr}.  For the history, see \LaTeX{} bug report 2646.
%
%   The check for \cs\ifvmode must not be part of a nested conditional.  \TeX{}
%   can't match \cs\if{}s with \cs\fi{}s properly when you nest tests for
%   \cs\if-type tokens.  See p.~211 of the \TeX{}book.
%
%   When we encounter an \cs\ifvmode, we must assume we are inside a
%   TextFontCommand declared by one of the two kernel versions mentioned above.
%   If not, we are in an unknown situation and we will bomb.  Since the error
%   message in this case won't be helpful, we warn the user in the log file.
%   We use scratch |d| to gobble both the \cs\ifvmode and what we expect will
%   follow the \cs\ifvmode, namely \code{\else\check@icr\fi}.  After swallowing
%   those, we reissue those same commands and then proceed with our lookahead.
%   We want to issue those commands, which conditionally introduce an italic
%   correction, before looking further ahead.
%    \begin{macrocode}
\newcommand\ti@q@handle@ifvmode {%
%  \DTypeout{Handling ifvmode}%
  \FrankenInfo{titles}
    {Handling an \string\ifvmode\space following a title.\MessageBreak
     If you now get an error that \string\sc@t@d\space does not\MessageBreak
     match its definition, this \string\ifvmode\space is\MessageBreak
     unexpected}%
  \DefName{sc@t@d} {\ifvmode\else\check@icr\fi} {%
    \ifvmode
    \else
      \check@icr
    \fi
    \aftergroup\ti@q@check
  }%
  \sc@t@d
}
%    \end{macrocode}
% \end{macro}
%
% \begin{macro}{\ti@q@handle@space}
%   Handle the case of a following space: gobble the space and call
%   \cname{ti@q@check}.
%
%   This little bit of trickery sneaks a space in as the \cs\def template,
%   thereby causing a space following \cname{ti@q@handle@space} to get gobbled.
%   We use the control character \cname{:} and restore its value.
%    \begin{macrocode}
\ReserveCS\ti@q@handle@space
\let\sc@t@a\:
\def\:{\ti@q@handle@space} \expandafter\def\: {\ti@q@check}
\let\:\sc@t@a
%    \end{macrocode}
%  \end{macro}
%
% \begin{macro}{\ti@q@handle@single@suck}
% \begin{macro}{\ti@q@handle@double@suck}
%   Handle the single and double sucking cases: gobble the closequotes token
%   with a \cs\def template, add some stuff to scratch |f| and call
%   \cname{ti@q@check}.  These are put in their own macros only to avoid
%   clutter above.
%    \begin{macrocode}
\newcommand*\ti@q@handle@double@suck [1] {%
%  \DTypeout{handle double suck: gobbling [\meaning#1]}%
%  \DTypeout{scratch f before: [\meaning\sc@t@f]}%
  \g@addto@macro\sc@t@f {\,\textquotedblright}%
%  \DTypeout{scratch f after: [\meaning\sc@t@f]}%
  \ti@q@check
}
\newcommand*\ti@q@handle@single@suck [1] {%
%  \DTypeout{handle single suck: gobbling [\meaning#1]}%
%  \DTypeout{scratch f before: [\meaning\sc@t@f]}%
  \g@addto@macro\sc@t@f {\,\textquoteright}%
%  \DTypeout{scratch f after: [\meaning\sc@t@f]}%
  \ti@q@check
}
%    \end{macrocode}
%  \end{macro}
%  \end{macro}
%
% \begin{macro}{\ti@q@handle@single@nosuck}
% \begin{macro}{\ti@q@handle@double@nosuck}
%   Handle the single and double nosucking cases.  Add inter-quote space to
%   scratch |f| and exit \cname{ti@q@@ifnch} with |true| or |false| depending
%   on whether we were looking for it.  We had to do it this way instead of the
%   normal \cname{if} test above at the end of \cname{ti@q@@ifnch} because
%   \cname{ti@close@double@nosuck} is more than one character long.
%    \begin{macrocode}
\newcommand*\ti@q@handle@double@nosuck [1] {%
% \DTypeout{handle double nosuck: gobbling [\meaning#1]}%
  \if@look@nosuck@
%      \DTypeout{And we're looking for \string\ti@close@double@nosuck.}%
      \g@addto@macro\sc@t@f {\,\textquotedblright}%
%      \DTypeout{After adding padding, F=[\meaning\sc@t@f]}%
      \let\sc@t@d\sc@t@a % the ``true'' clause of ti@q@ifnextcharin
  \else
%    \DTypeout{But we're not looking for \string\ti@close@double@nosuck.}%
%    \DTypeout{F is unchanged, F=[\meaning\sc@t@f]}%
    \let\sc@t@d\sc@t@b % the ``false'' clause of ti@q@ifnextcharin
  \fi
  \ti@q@check
}
\newcommand*\ti@q@handle@single@nosuck [1] {%
% \DTypeout{handle single nosuck: gobbling [\meaning#1]}%
  \if@look@nosuck@
%      \DTypeout{And we're looking for \string\ti@close@single@nosuck.}%
      \g@addto@macro\sc@t@f {\,\textquoteright}%
%      \DTypeout{After adding padding, F=[\meaning\sc@t@f]}%
      \let\sc@t@d\sc@t@a % the ``true'' clause of ti@q@ifnextcharin
  \else
%    \DTypeout{But we're not looking for \string\ti@close@single@nosuck.}%
%    \DTypeout{F is unchanged, F=[\meaning\sc@t@f]}%
    \let\sc@t@d\sc@t@b % the ``false'' clause of ti@q@ifnextcharin
  \fi
  \ti@q@check
}
%    \end{macrocode}
%  \end{macro}
%  \end{macro}
%
% \section{Words and phrases}
%
% \begin{macro}{\word}
% \begin{macro}{\foreign}
% \begin{macro}{\foreignword}
% \begin{macro}{\phrase}
% \begin{macro}{\term}
% \begin{macro}{\defn}
%   \mbox{}
%    \begin{macrocode}
\newlet\word\textitswitch
\newlet\foreign\textitswitch
%    \end{macrocode}
% \todo{\cs\phrase is the result of expansion here: what effect will this have
% on its proper nesting, and is this something to worry about?}
%    \begin{macrocode}
\newcommand\foreignword [1] {%
  \phrase{\word{#1}}%
}
%    \end{macrocode}
% The \cs\@ cancels the effect on spacing of any final punctuation in the
% argument.
%    \begin{macrocode}
% \newlet\phrase\WrapquotesSK -- whoops, doesn't work as intended,
% \phrase{foo}s puts a space before the following `s'

% old definition:
\newcommand\phrase [1] {%
  \textquoteleft #1\textquoteright\@%
}
\newlet\term\textitswitch
\newlet\defn\textslswitch
%    \end{macrocode}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
%
% \section{Titles}
%
% \begin{macro}{\book}
% \begin{macro}{\journal}
% \begin{macro}{\music}
% \begin{macro}{\article}
% \begin{macro}{\storytitle}
% \begin{macro}{\poemtitle}
% \begin{macro}{\play}
% \begin{macro}{\craft}
% \begin{macro}{\species}
%   \mbox{}
%    \begin{macrocode}
\newlet\book\textitswitch
\newlet\journal\textitswitch
\newlet\music\textitswitch
\newlet\article\Wrapquotes
\newlet\storytitle\Wrapquotes
\newlet\poemtitle\Wrapquotes
\newlet\play\textitswitch % \manualref{7.145}
\newlet\craft\textitswitch
\newlet\species\textitswitch
%    \end{macrocode}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
% \end{macro}
%
% \part{Configuration}
%
% User alterations and additions and package testing are in a configuration
% file.
%    \begin{macrocode}
\InputIfFileExists{titles.cfg}{}{}
%    \end{macrocode}
%
%  The contents of the \emph{distributed} configuration file are below.
%
% \InputIfFileExists{titles.cfg}{}{}
%
% \Finale